Nucleoside Tetra- and Pentaphosphates Prepared Using a Tetraphosphorylation Reagent Are Potent Inhibitors of Ribonuclease A

The MIT Faculty has made this article openly available. Please share how this access benefits you. Your story matters.

Citation Shepard, Scott M. et al. "Nucleoside Tetra- and Pentaphosphates Prepared Using a Tetraphosphorylation Reagent Are Potent Inhibitors of Ribonuclease A." Journal of the American Chemical Society 141, 46 (October 2019): 18400–18404 © 2019 American Chemical Society

As Published http://dx.doi.org/10.1021/jacs.9b09760

Publisher American Chemical Society (ACS)

Version Author's final manuscript

Citable link https://hdl.handle.net/1721.1/128544

Terms of Use Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. HHS Public Access Author manuscript

Author ManuscriptAuthor Manuscript Author J Am Chem Manuscript Author Soc. Author Manuscript Author manuscript; available in PMC 2020 February 12. Published in final edited form as: J Am Chem Soc. 2019 November 20; 141(46): 18400–18404. doi:10.1021/jacs.9b09760.

Nucleoside Tetra- and Pentaphosphates Prepared Using a Tetraphosphorylation Reagent Are Potent Inhibitors of Ribonuclease A

Scott M. Shepard†, Ian W. Windsor†, Ronald T. Raines*, Christopher C. Cummins* Department of Chemistry, Massachusetts Institute of Technology, Cambridge Massachusetts 02139, United States

Abstract Adenosine and uridine 5′-tetra- and 5′-pentaphosphates were synthesized from an activated tetrametaphosphate ([PPN]2[P4O11], [PPN]2[1], PPN = bis(triphenylphosphine)iminium) and subsequently tested for inhibition of the enzymatic activity of ribonuclease A (RNase A). Reagent [PPN]2[1] reacts with unprotected uridine and adenosine in the presence of a base under anhydrous conditions to give nucleoside tetrametaphosphates. Ring opening of these intermediates with tetrabutylammonium hydroxide ([TBA][OH]) yields adenosine and uridine tetraphosphates (p4A, p4U) in 92% and 85% yields, respectively, from the starting nucleoside. Treatment of ([PPN]2[1]) with AMP or UMP yields nucleoside-monophosphate tetrametaphosphates (cp4pA, cp4pU) having limited aqueous stability. Ring opening of these ultraphosphates with [TBA][OH] yields p5A and p5U in 58% and 70% yield from AMP and UMP, respectively. We characterized inorganic and nucleoside-conjugated linear and cyclic oligophosphates as competitive inhibitors of RNase A. Increasing the chain length in both linear and cyclic inorganic oligophosphates resulted in improved binding affinity. Increasing the length of oligophosphates on the 5′ position of adenosine beyond three had a deleterious effect on binding. Conversely, uridine nucleotides bearing 5′ oligophosphates saw progressive increases in binding with chain length. We solved X- ray cocrystal structures of the highest affinity binders from several classes. The terminal phosphate of p5A binds in the P1 enzymic subsite and forces the oligophosphate to adopt a convoluted conformation, while the oligophosphate of p5U binds in several extended conformations, targeting multiple cationic regions of the active-site cleft.

Secretory ribonucleases (RNases) are a diverse family of enzymes that catalyze the cleavage of RNA to elicit biological functions ranging from cell signaling to innate immunity.1,2

*Corresponding Authors: [email protected]; [email protected]. †S.M.S. and I.W.W. contributed equally. Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.9b09760. Synthetic details, spectra, kinetic data, and crystallographic data collection and refinement statistics (PDF) Accession Codes Structure data for the new compounds are available from the Protein Database under the following PDB codes: RNase A·cP6i complex, 6pvu; RNase A·p5A complex, 6pvv; RNase A·cp4pA complex, 6pvw; RNase A·p5U complex, 6pvx. The authors declare the following competing financial interest(s): The tetraphosphorylation reagent is covered in patent US10017388B2. Shepard et al. Page 2

Fundamental knowledge generated by studying RNase A, which derives from the bovine Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author pancreas, has shaped the fields of enzymology and protein chemistry.3,4 Furthermore, mammalian RNases have been shown to have angiogenic5 and neurotoxic activities,6 and targeted inhibitors of these enzymes may have human therapeutic potential.7 RNase A binds its substrates in enzymic subsites that interact with phosphoryl groups and nucleobases (Figure 1).8,9

Atypical nucleotides are among the best small-molecule inhibitors of RNase A. Diadenosine oligophosphates (Table 1, entries 5–7) are micromolar to submicromolar inhibitors that exhibit increasing affinity with longer phosphate chain lengths.10 Additionally, the highest affinity small-molecule inhibitors of RNase A, pyrophosphate-linked dinucleotides (Table 1, entries 1–4), have enhanced inhibition activity upon further phosphorylation.11 These observations prompted us to ask: can a simple oligophosphate on its own or appended to a single nucleoside serve as an effective small-molecule inhibitor of RNase A?

In addition to canonical nucleoside mono-, di-, and triphosphates, nucleosides bearing longer oligophosphate chains are potent signaling molecules in biology.14–18 These and other related morphologies, such as dinucleotide oligophosphates,19–21 have been implicated in a variety of biological processes and ailments including hypertension22 and bacterial accumulation of polyphosphate.23–26 Nonetheless, the synthesis of oligophosphorylated compounds typically increases in difficulty with longer phosphate chains. Recently, methods have been developed to efficiently couple a triphosphate chain in one operation from 3− 14,27–32 trimetaphosphate (P3O9 ). Here, we extend this methodology to the 4− tetraphosphorylation of biomolecules, utilizing tetrametaphosphate (P4O12 ) to synthesize nucleoside tetraphosphates (p4N) and nucleoside pentaphosphates (p5N). Previous syntheses 33 of p4N have suffered from extremely low yields, requiring nucleoside triphosphates as starting materials,34 or iterative syntheses to add each additional phosphoryl group.13,23 The state-of-the-art synthesis of p4N involves coupling of trimetaphosphate and nucleoside 14 monophosphates. Here, we describe the facile synthesis of p4N and p5N by coupling tetrametaphosphate with nucleosides and nucleoside monophosphates, respectively (Figure 2). Utilizing tetraphosphorylation reagent 1 permits unprotected nucleosides to be converted to the corresponding p4N efficiently in a single operation. This is in contrast to the method of Taylor, which requires nucleoside monophosphates as the substrate.14 While enzymatic 35–37 methods have been reported, the few reported chemical syntheses of p5N have been limited in scope and low yielding.23,33

The activated tetrametaphosphate, [PPN]2[1], is synthesized by protonation of tetrametaphosphate and subsequent dehydration.38 Treatment of adenosine or uridine with [PPN]2[1] under rigorously anhydrous conditions leads to selective phosphorylation of the 5′ position. No satisfactory purification could be found for the resulting nucleoside- substituted tetrametaphosphates (cp4N, Figure 3), but treatment with [TBA][OH] results in ring opening to the linear tetraphosphates. HPLC purification in triethylammonium acetate buffer of the resulting mixtures gives adenosine tetraphosphate (p4A, Table 1, entry 18, 93% yield) and uridine tetraphosphate (p4U, Table 1, entry 25, 85% yield) as pure triethylammonium salts.

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 3

Nucleoside 5′-pentaphosphates were obtained similarly by treatment of [PPN]2[1] with the Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author anhydrous TBA salts of pA and pU. The intermediate nucleoside-monophosphate substituted tetrametaphosphates, cp4pN, could be isolated in reasonable purity and were found to be stable in aqueous solution for several hours at room temperature before hydrolyzing to a mixture of nucleoside monophosphate, tetrametaphosphate, and nucleoside pentaphosphate. Treatment of cp4pN with excess [TBA][OH] results in selective ring opening to p5N in 24 h. The products were again purified by HPLC in triethylammonium acetate buffer, providing p5A (Table 1, entry 19, 58% yield) and p5U (Table 1, entry 26, 70% yield) as triethylammonium salts (Figure 3).

Reagent 1 is highly moisture sensitive and must be prepared, stored, and utilized in an anhydrous environment, ideally inside a glovebox. We therefore developed a second phosphorylation methodology for the syntheses of both p4N and p5N, activating 38 [PPN]2[P4O12H2] in situ with dicyclohexylcarbodiimide (DCC) to form reagent 1. These reagents are bench stable, and this methodology can be utilized conveniently with a Schlenk line, although it suffers from lower yields (SI Sections 2.2 and 2.5).

We performed inhibition kinetics using a fluorogenic substrate as described previously39 to assess the binding of oligophosphates to RNase A. In addition to the synthesized molecules, p4N and p5N, we assessed inhibition kinetics for a variety of inorganic phosphates to evaluate our hypothesis that longer oligophosphate chains increase binding affinity. Complementing previous reports of weak RNase A inhibition by orthophosphate (Pi, Table 1, entry 8) and pyrophosphate (P2i, Table 1, entry 9), we tested P3i and P4i (Table 1, entries 10–11). The measured Ki values decrease for longer phosphates with a value of 23 μM for P4i. Although a longer inorganic oligophosphate may be a more potent inhibitor, each subsequent phosphoryl group has a diminished impact on lowering the Ki value. We similarly tested inorganic tri- (cP3i, Table 1, entry 12), tetra- (cP4i, Table 1, entry 13), and hexametaphosphate (cP6i, Table 1, entry 14) as these cyclic phosphates have been largely ignored in biological systems despite their indefinite stabilities near neutral pH. A similar trend was observed with increased inhibition of RNase A for longer oligophosphates and diminishing returns for each additional phosphoryl group. The metaphosphates are modestly less effective inhibitors than the corresponding linear phosphate.

Inhibition of RNase A by adenosine nucleotides does not follow the same simple trend as inorganic phosphates. Reported Ki values are given in Table 1, entries 15–19 for adenosine 5′-oligophosphates ranging from monophosphate (pA) to pentaphosphate (p5A). Inhibition increases from pA to the strongest inhibitor of this series, p3A with a Ki value of 0.86 μM. p4A and p5A are somewhat less effective inhibitors with Ki values of 2.1 and 1.4 μM, respectively, indicating that the role of the oligophosphate chain in binding is not reducible simply to a Coulombic interaction. Furthermore, we tested the hydrolytically sensitive ultraphosphate cp4pA and found that it was a superior inhibitor with a Ki value of 0.48 μM (Table 1, entry 20), suggesting that this unusual phosphate geometry is better able to target the active site.

In contrast to 5′-adenosine nucleotides, RNase A inhibition by 5′-uridine nucleotides follows a simple trend analogous to inorganic phosphates. In the series pU to p5U (Table 1,

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 4

entries 22–26), inhibition increases successively with longer oligophosphate chains, and the Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author strongest inhibitor is p5U with a Ki value of 0.068 μM. This mononucleotide approaches the potency of the best pyrophosphate-linked dinucleotide inhibitors of RNase A (Table 1, 11 entries 2 and 4). The ultraphosphate species cp4pU is, however, a weaker inhibitor with a Ki value of 0.98 μM (Table 1, entry 27). Finally, we assessed the salt dependence of + inhibition by p5U. Compared to a DNA tetramer AUAA, which exhibits a δlogKd/δlog[Na ] 8 of 2.3, p5U has a greater value of 3.2 (Figure S30), indicating a greater dependence on Coulombic interactions. Nonetheless, p5U is among the most potent small-molecule inhibitors reported of RNase A and maintains high affinity at physiological salt concentrations.

To understand the differing trends in inhibition of RNase A by adenosine and uridine nucleotides, we solved X-ray cocrystal structures of several of these ligands bound to RNase A. The structure of bound cP6i, which is the first protein crystal structure to contain a metaphosphate, shows the ligand bound primarily in the P1 subsite (Figure S31A). This is 44 45 analogous to structures of bound Pi, (5rsa) and P2i (2w5m), suggesting that the binding of these inorganic phosphates is largely conserved. Differences in inhibition are attributable to Coulombic interactions and slight variations in hydrogen bonding.

The enhanced inhibition of p3A over other adenosine nucleotides is illuminated by 45 crystallography. In structures of both p3A (2w5g, Figure 4A) and p5A (6pvv, Figure 4B), the terminal phosphoryl group of the oligophosphate chain binds in the P1 subsite with hydrogen bonds to His12 and His119. For the longer oligophosphate in p5A to target the same site requires the phosphate chain to form a loop. Apparently, the thermodynamic penalty associated with this constrained geometry is sufficient to overcome the greater Coulombic attraction between the enzyme and the more highly charged pentaphosphate, resulting in weaker inhibition. Furthermore, the strong inhibition by cp4pA is attributable in part to more efficient targeting of Lys7 in this unusual phosphate geometry (6pvw, Figure 4C).

In 5′-uridine nucleotides, longer oligophosphate chains are not detrimental to inhibition of RNase A. Pyrimidine nucleobases are preferred in the B1 subsite, but a previous structure of p2U revealed occupancy of both the P0/B1 and P1/B2 subsites by a pair of these ligands (3dxh, Figure 4D). Chain A of the structure of p5U bound to RNase A (6pvx, Figure 4E) shares this mode of recognition; however, only a single p5U binds the P1/B1 subsites in chain B (Figure 4F). In both chains, the oligophosphates extend well beyond the phosphoryl group-binding subsites and can efficiently target several cationic regions of the active-site cleft. Thermodynamically unfavorable constraints are not imposed on the polyphosphate chain; therefore, longer, more highly charged oligophosphates could improve affinity of 5′- uridine nucleotides.

The role of polyphosphates in biological systems has come under increasing study. Here, we contribute to the synthetic methodology to create these species as well as perform analyses of their function. The activated tetrametaphosphate reagent [PPN]2[1] is a useful synthetic tool for synthesizing polyphosphate chains of four units or longer by reaction with suitable anhydrous nucleophiles. Furthermore, intermediates containing substituted metaphosphates

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 5

can, in some cases, be isolated as pure compounds that possess modest aqueous stability. We Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author showed that binding of polyphosphorylated compounds to the active-site cleft of RNase A generally follows the simple trend that affinity increases with oligophosphate length, with 5′-adenosine oligophosphates being the notable exception. The limit of affinity enhancement conferred by lengthening the phosphate chain of 5′-uridine nucleotides remains to be determined, and we are working toward synthesizing p6U as well as exploring phosphorylation on 3′-nucleotide positions. Furthermore, this work presents the first crystal structures of a metaphosphate (Figure S31A) or an ultraphosphate (Figure 4C) bound to a protein. This demonstrates that complex polyphosphate morphologies that have been largely excluded from consideration in aqueous media may in fact be relevant to biological systems, opening new avenues for biochemical studies and drug development.

Supplementary Material

Refer to Web version on PubMed Central for supplementary material.

ACKNOWLEDGMENTS

This work was supported by the NIH under Grant No. R01 CA073808.

REFERENCES (1). Lu L; Li J; Moussaoui M; Boix E Immune modulation by human secreted RNases at the extracellular space. Front. Immunol 2018, 9, 1012. [PubMed: 29867984] (2). Sheng J; Xu Z Three decades of research on angiogenin: a review and perspective. Acta Biochim. Biophys. Sin 2016, 48, 399–410. [PubMed: 26705141] (3). Cuchillo CM; Nogués MV; Raines RT Bovine pancreatic ribonuclease: fifty years of the first enzymatic reaction mechanism. Biochemistry 2011, 50, 7835–7841. [PubMed: 21838247] (4). Marshall GR; Feng JA; Kuster DJ Back to the future: ribonuclease A. Biopolymers 2008, 90, 259– 277. [PubMed: 17868092] (5). Olson KA; Fett JW; French TC; Key ME; Vallee BL Angiogenin antagonists prevent tumor growth in vivo. Proc. Natl. Acad. Sci. U. S. A 1995, 92, 442–446. [PubMed: 7831307] (6). Gleich GJ; Loegering DA; Bell MP; Checkel JL; Ackerman SJ; McKean DJ Biochemical and functional similarities between human eosinophil-derived neurotoxin and eosinophil cationic protein: homology with ribonuclease. Proc. Natl. Acad. Sci. U. S. A 1986, 83, 3146–3150. [PubMed: 3458170] (7). Russo N; Shapiro R Potent inhibition of mammalian ribonucleases by 3′,5′-pyrophosphate-linked nucleotides. J. Biol. Chem 1999, 274, 14902–14908. [PubMed: 10329690] (8). Fisher BM; Ha J-H; Raines RT Coulombic forces in protein-RNA interactions: binding and cleavage by ribonuclease A and variants at Lys7, Arg10, and Lys66. Biochemistry 1998, 37, 12121–12132. [PubMed: 9724524] (9). Fontecilla-Camps JC; de Llorens R; Le Du M; Cuchillo CM Crystal structure of ribonuclease A·d(ApTpApApG) complex. Direct evidence for extended substrate recognition. J. Biol. Chem 1994, 269, 21526–21531. [PubMed: 8063789] (10). Kumar K; Jenkins JL; Jardine AM; Shapiro R Inhibition of mammalian ribonucleases by endogenous adenosine dinucleotides. Biochem. Biophys. Res. Commun 2003, 300, 81–86. [PubMed: 12480524] (11). Leonidas DD; Shapiro R; Irons LI; Russo N; Acharya KR Toward rational design of ribonuclease inhibitors: High-resolution crystal structure of a ribonuclease A complex with a potent 3′,5′- pyrophosphate-linked dinucleotide inhibitor. Biochemistry 1999, 38, 10287–10297. [PubMed: 10441122]

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 6

(12). Fisher BM; Grilley JE; Raines RT A new remote subsite in ribonuclease A. J. Biol. Chem 1998, Author ManuscriptAuthor Manuscript Author Manuscript Author 273, 34134–34138. Manuscript Author [PubMed: 9852072] (13). Strenkowska M; Wanat P; Ziemniak M; Jemielity J; Kowalska J Preparation of synthetically challenging nucleotides using cyanoethyl P-imidazolides and microwaves. Org. Lett 2012, 14, 4782–4785. [PubMed: 22966945] (14). Mohamady S; Taylor SD Synthesis of nucleoside tetraphosphates and dinucleoside pentaphosphates via activation of cyclic trimetaphosphate. Org. Lett 2013, 15, 2612–2615. [PubMed: 23668391] (15). R. Kore A; Yang B; Srinivasan B; Conrad R Chemical and enzymatic synthesis of nucleoside tetraphosphates. Curr. Org. Chem 2014, 18, 1621–1650. (16). Jankowski V; Tölle M; Vanholder R; Schönfelder G; van der Giet M; Henning L; Schlüter H; Paul M; Zidek W; Jankowski J Uridine adenosine tetraphosphate: a novel endothelium-derived vasoconstrictive factor. Nat. Med 2005, 11, 223. [PubMed: 15665829] (17). Han Q; Gaffney BL; Jones RA One-flask synthesis of dinucleoside tetra and pentaphosphates. Org. Lett 2006, 8, 2075–2077. [PubMed: 16671785] (18). Sundaralingam M Stereochemistry of nucleic acids and their constituents. IV. Allowed and preferred conformations of nucleosides, nucleoside mono-, di-, tri-, tetraphosphates, nucleic acids and polynucleotides. Biopolymers 1969, 7, 821–860. (19). Jovanovic A; Jovanovic S; Mays DC; Lipsky JJ; Terzic A Diadenosine 5′,5″,-P1,P5- pentaphosphate harbors the properties of a signaling molecule in the heart. FEBS Lett. 1998, 423, 314–318. [PubMed: 9515730] (20). Luo J; Jankowski V; Gungar N; Neumann J; Schmitz W; Zidek W; Schlüter H; Jankowski J Endogenous diadenosine tetraphosphate, diadenosine pentaphosphate, and diadenosine hexaphosphate in human myocardial tissue. Hypertension 2004, 43, 1055–1059. [PubMed: 15066958] (21). Miras-Portugal MT; Gualix J; Pintor J The neurotransmitter role of diadenosine polyphosphates. FEBS Lett. 1998, 430, 78–82. [PubMed: 9678598] (22). Matsumoto T; Goulopoulou S; Taguchi K; Tostes RC; Kobayashi T Constrictor prostanoids and uridine adenosine tetraphosphate: vascular mediators and therapeutic targets in hypertension and diabetes. Br. J. Pharmacol 2015, 172, 3980–4001. [PubMed: 26031319] (23). Mordhorst S; Singh J; Mohr MK; Hinkelmann R; Keppler M; Jessen HJ; Andexer JN Several polyphosphate kinase 2 enzymes catalyse the production of adenosine 5′-polyphosphates. ChemBioChem 2019, 20, 1019–1022. [PubMed: 30549179] (24). Bhandari R; Saiardi A; Ahmadibeni Y; Snowman AM; Resnick AC; Kristiansen TZ; Molina H; Pandey A; Werner JK; Juluri KR; Xu Y; Prestwich GD; Parang K; Snyder SH Protein pyrophosphorylation by inositol pyrophosphates is a posttranslational event. Proc. Natl. Acad. Sci. U. S. A 2007, 104, 15305–15310. [PubMed: 17873058] (25). Marmelstein AM; Yates LM; Conway JH; Fiedler D Chemical pyrophosphorylation of functionally diverse peptides. J. Am. Chem. Soc 2014, 136, 108–111. [PubMed: 24350643] (26). Azevedo C; Livermore T; Saiardi A Protein polyphosphorylation of lysine residues by inorganic polyphosphate. Mol. Cell 2015, 58, 71–82. [PubMed: 25773596] (27). Shepard SM; Cummins CC Functionalization of intact trimetaphosphate: a triphosphorylating reagent for C, N, and O nucleophiles. J. Am. Chem. Soc 2019, 141, 1852–1856. [PubMed: 30646689] (28). Mohamady S; Taylor SD Synthesis of nucleoside triphosphates from 20–30-protected nucleosides using trimetaphosphate. Org. Lett 2016, 18, 580–583. [PubMed: 26759914] (29). Mohamady S; Taylor SD Synthesis of nucleoside 5′tetraphosphates containing terminal fluorescent labels via activated cyclic trimetaphosphate. J. Org. Chem 2014, 79, 2308–2313. [PubMed: 24552623] (30). Singh J; Steck N; De D; Hofer A; Ripp A; Captain I; Keller M; Wender PA; Bhandari R; Jessen HJ A phosphoramidite analogue of cyclotriphosphate enables iterative polyphosphorylations. Angew. Chem. Int. Ed 2019, 58, 3928–3933.

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 7

(31). Fernandes-Cunha GM; McKinlay CJ; Vargas JR; Jessen HJ; Waymouth RM; Wender PA Author ManuscriptAuthor Manuscript Author Manuscript Author Delivery of inorganic Manuscript Author polyphosphate into cells using amphipathic oligocarbonate transporters. ACS Cent. Sci 2018, 4, 1394–1402. [PubMed: 30410977] (32). Azevedo C; Singh J; Steck N; Hofer A; Ruiz FA; Singh T; Jessen HJ; Saiardi A Screening a protein array with synthetic biotinylated inorganic polyphosphate to define the human polyP- ome. ACS Chem. Biol 2018, 13, 1958–1963. [PubMed: 29924597] (33). Ko H; Carter RL; Cosyn L; Petrelli R; de Castro S; Besada P; Zhou Y; Cappellacci L; Franchetti P; Grifantini M; Calenbergh SV; Harden TK; Jacobson KA Synthesis and potency of novel uracil nucleotides and derivatives as P2Y2 and P2Y6 receptor agonists. Bioorg. Med. Chem 2008, 16, 6319–6332. [PubMed: 18514530] (34). Zuberek J; Jemielity J; Jablonowska A; Stepinski J; Dadlez M; Stolarski R; Darzynkiewicz E Influence of electric charge variation at residues 209 and 159 on the interaction of eIF4E with the mRNA 5′ terminus. Biochemistry 2004, 43, 5370–5379. [PubMed: 15122903] (35). Guranowski A; Sillero MG; Sillero A Adenosine 5′tetraphosphate and adenosine 5′- pentaphosphate are synthesized by yeast acetyl coenzyme A synthetase. J. Bacteriol 1994, 176, 2986–2990. [PubMed: 7910605] (36). Guranowski A; de Diego A; Sillero A; Günther Sillero MA Uridine 5′-polyphosphates (p4U and p5U) and uridine(5′)-polyphospho(5′)nucleosides (UpnNs) can be synthesized by UTP: glucose-1-phosphate uridylyltransferase from Saccharomyces cerevisiae. FEBS Lett. 2004, 561, 83–88. [PubMed: 15013755] (37). Nakajima H; Tomioka I; Kitabatake S; Dombou M; Tomita K Facile and selective synthesis of diadenosine polyphosphates through catalysis by leucyl t-RNA synthetase coupled with ATP regeneration. Agric. Biol. Chem 1989, 53, 615–623. (38). Jiang Y; Chakarawet K; Kohout AL; Nava M; Marino N; Cummins CC Dihydrogen 2− tetrametaphosphate, [P4O12H2] : synthesis, solubilization in organic media, preparation of its 2− anhydride [P4O11] and acidic methyl ester, and conversion to tetrametaphosphate metal complexes via protonolysis. J. Am. Chem. Soc 2014, 136, 11894–11897. [PubMed: 25102033] (39). Kelemen BR; Klink TA; Behike MA; Eubanks SR; Leland PA; Raines RT Hypersensitive substrate for ribonucleases. Nucleic Acids Res. 1999, 27, 3696–3701. [PubMed: 10471739] (40). Anderson DG; Hammes GG; Walz FG Binding of phosphate ligands to ribonuclease A. Biochemistry 1968, 7, 1637–1645. [PubMed: 4297049] (41). Russo N; Shapiro R; Vallee BL 5′-Diphosphoadenosine 3′phosphate is a potent inhibitor of bovine pancreatic ribonuclease A. Biochem. Biophys. Res. Commun 1997, 231, 671–674. [PubMed: 9070868] (42). Tsirkone VG; Dossi K; Drakou C; Zographos SE; Kontou M; Leonidas DD Inhibitor design for ribonuclease A: the binding of two 5′-phosphate uridine analogues. Acta Crystallogr., Sect. F: Struct. Biol. Cryst. Commun 2009, 65, 671–677. (43). Dossi K; Tsirkone VG; Hayes JM; Matoušek J; Poučková P; Souček J; Zadinova M; Zographos SE; Leonidas DD Mapping the ribonucleolytic active site of bovine seminal ribonuclease. The binding of pyrimidinyl phosphonucleotide inhibitors. Eur. J. Med. Chem 2009, 44, 4496–4508. [PubMed: 19643512] (44). Wlodawer A; Borkakoti N; Moss D; Howlin B Comparison of two independently refined models of ribonuclease-A. Acta Crystallogr., Sect. B: Struct. Sci 1986, 42, 379–387. (45). Holloway DE; Chavali GB; Leonidas DD; Baker MD; Acharya KR Influence of naturally- occurring 5′-pyrophosphate-linked substituents on the binding of adenylic inhibitors to ribonuclease A: an X-ray crystallographic study. Biopolymers 2009, 91, 995–1008. [PubMed: 19191310]

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 8 Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author

Figure 1. Cocrystal structure of RNase A bound to an AUAA DNA tetramer revealed the subsites that recognize nucleobases and phosphoryl groups (1rcn, top). The mainchain of RNase A is traced with a cartoon, key active-site and cysteine residues are shown as sticks, and ligands are shown as balls-and-sticks. Residues in subsites are colored blue (P2), red (P1), and green (P0). A cartoon representation of the RNase A active site showing the preferred binder for 12 each subsite (bottom). For simplicity, the P−1 subsite is not shown.

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 9 Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author

Figure 2. Synthesis of nucleoside tetraphosphates by Kowalska13 and Taylor14 compared to this synthesis of nucleoside tetra- and pentaphosphates.

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 10 Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author

Figure 3. (A) i. [PPN]2[1] (1.5 equiv) and triethylamine (2 equiv) (DMF, N2 atmosphere, 48 h); ii. [TBA][OH] (7.5 equiv) (DMF/H2O, 2 h) followed by HPLC (50 mM triethylammonium acetate (TEAA)). (B) i. [PPN]2[1] (1.1 equiv) (DMF, N2 atmosphere, 30 min); ii. [TBA] [OH] (4.5 equiv) (DMF/H2O, 24 h) followed by HPLC (50 mM TEAA).

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 11 Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author

Figure 4. Interactions between nucleotides and the active site of RNase A. The structures are depicted as described in Figure 1. (A) pA and P3A bind the active site by positioning the terminal phosphate group in the P1 subsite with the adenosine base in the B2 subsite (1z6s, 2w5g). (B) p5A binds similarly to shorter adenosine nucleotides (6pvv). (C) cp4pA binds similarly to p5A, but more efficiently targets Lys7 of the P2 subsite (6pvw). (D) Two molecules of p2U bind RNase A; however, only the B1 subsite is efficiently targeted (3dxh). (E) Chain A

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 12

of the RNase A·p5U complex is similar to p2U (6pvx). (F) In chain B, p5U only binds in the Author ManuscriptAuthor Manuscript Author Manuscript Author Manuscript Author B1 subsite and alternatively targets the P1 and P2 subsites (6pvx).

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12. Shepard et al. Page 13

Table 1.

Author ManuscriptAuthor Inhibition Manuscript Author Constants of Manuscript Author Inorganic Phosphates Manuscript Author and Nucleotides for RNase A

a entry compound Ki (μM) ref

1 dUppAp 0.12 7 2 pdUppAp 0.027 7 3 TppdA 4 7 4 pdTppAp 0.041 7

5 Ap3A 29 10

6 Ap4A 2.6 10

7 Ap5A 0.23 10

8 Pi 4600 40

9 P2i 170 40

10 P3i 23 ± 1 this work

11 P4i 6.8 ± 0.2 this work

12 cP3i 960 ± 80 this work

13 cP4i 30 ± 0.8 this work

14 cP6i 6.2 ± 0.1 this work 15 pA 170 ± 6 this work

16 p2A 1.2 41

17 p3A 0.86 10

18 p4A 2.1 ± 0.2 this work

19 p5A 1.4 ± 0.06 this work

20 cp4pA 0.48 ± 0.03 this work 21 ppAp 0.24 7 22 pU 4000 42

23 p2U 650 43

24 p3U 8.3 ± 0.3 this work

25 p4U 1.8 ± 0.1 this work

26 p5U 0.068 ± 0.007 this work

27 cp4pU 0.98 ± 0.07 this work

a Values from this work are reported ± the standard error of fitting one-site binding equation.

J Am Chem Soc. Author manuscript; available in PMC 2020 February 12.